Free transcription · runs in your browser
Speech to text that stays with you.
Murmurhouse turns recordings and videos into text and subtitles on your own device. No upload, no account, no minute limits, and it works offline after the first use.
Three steps, all on your device
Your computer does the listening. Nothing you drop in is sent anywhere.
Drop a file or record
MP3, M4A, WAV, MP4, MOV, WebM and more. Or press record and talk.
Your device transcribes
OpenAI’s open Whisper model runs in your browser, in around 100 languages, including Luxembourgish.
Fix, copy, download
Click any line to correct it. Copy the text, or download TXT, SRT or VTT subtitles.
Made for recordings you’d rather not upload
Interviews, therapy and coaching notes, client calls, lectures, voice memos, family stories. Most transcription sites send your audio to their servers. Murmurhouse doesn’t have servers that could receive it.
Private by design
No upload step exists. Your transcripts are saved only in your browser, and you can delete them anytime.
Any computer
Windows, Mac, Linux and Chromebook in Chrome, Edge, Firefox or Safari. Install it as an app if you like.
No limits, no ads
No minutes to count, no watermark, no sign-up. Long files just take a little longer.
Free to use. Pay once if you want more.
Everything you need to transcribe is free. Pro will be a one-time purchase, with no subscription.
Free
- Unlimited transcription with the Fast and Balanced models
- About 100 languages, plus translation to English
- TXT, SRT and VTT downloads with your corrections
- Record straight from your microphone
- Your 10 most recent transcripts saved in your browser
Pro · coming soon
- Everything in Free
- Accurate model for difficult audio and accents
- Batch: drop many files and transcribe them in a row
- Word and Markdown exports
- Keep 100 transcripts in your browser
One email on launch day, then we delete your address. Privacy
Questions
Is Murmurhouse really free?
Yes. Transcribing with the Fast and Balanced models is free, with no ads, no account and no limit on minutes. Pro will be an optional one-time purchase.
Is my audio uploaded anywhere?
No. The speech model runs inside your browser on your own device. Your recordings and transcripts never leave it. The only download is the speech model itself, which your browser fetches once from Hugging Face and then keeps.
Does it work offline?
Yes, after the first use. Once the model is saved in your browser, you can install Murmurhouse from your browser menu (“Install app”) and transcribe without a connection.
Which languages does it understand?
Around 100 languages, including English, French, German, Luxembourgish, Spanish, Italian, Portuguese, Dutch, Romanian and Polish. It can also translate speech from those languages into English text.
How long can a recording be?
There is no fixed limit. An hour of audio is fine on most laptops. Long files take a while because your own computer does the work, so keep the tab open until it finishes.
Which files can I use?
Most audio and video files your browser can play: MP3, M4A, WAV, OGG, Opus, FLAC, WebM, MP4 and MOV. You can also record straight from your microphone.
Can I make subtitles?
Yes. Download the transcript as SRT or VTT subtitles with timings, or as plain text. You can fix any words before you download.
How accurate is it?
It uses OpenAI’s open Whisper models. Clear speech in common languages comes out very well; noisy rooms, heavy accents and several people talking at once lead to more mistakes. You can correct any line by clicking it.
Try it with your next recording
Open Murmurhouse, drop in a file, and read along as it’s transcribed.